Jacques van Niekerk
Expertise as a (dis)Service
Updated: Sep 21, 2022
Every business must carefully consider whether they can afford to commit to a single cloud platform.
Amazon AWS recently launched AWS IQ - a service that seamlessly integrates the action of sourcing, managing and paying for AWS experts with your existing AWS accounts and billing structures.
AWS IQ verifies the skills of the people offering their services using AWS certifications and adds a 3% charge on top. Billing is through your existing AWS account.
It seems a natural step for Amazon to offer this type of service to complement the availability of professional services, partners and managed services. IQ is specifically aimed at the small and medium sized business.
AWS IQ is of course focused purely on Amazon AWS technology. There are many services that offer broader expertise as a service, including Talmix, Outvise and more.
Over the years we've seen an inevitable move from self-managed infrastructure to cloud based solutions. Except at the very largest scale, and for specialist applications, it rarely makes sense to acquire and manage your own hardware. Yet the services and support structure around cloud based infrastructure is, in a sense, becoming restrictive in nature. Once a decision has been made to adopt a specific vendor's services, there is a significant cost associated with architecting and implementing a platform that is easily transportable to different cloud providers.
The types of expertise that is built around a platform and that is offered as a service will be structured to advise the customer to use the platform to it's fullest capacity. This is true for all platforms - not only Amazon. In this sense such expertise may be a disservice to a business, because optimising for the use of a particular platform is most likely going to make it far more difficult to migrate to a different cloud provider.
We argue that every business should carefully consider whether they may have to move from one cloud provider to another (there are valid reasons for this - for example if you acquire customers who are competitors to a given vendor and cannot host their data on a particular cloud).
Best practise is to balance the cost of having a portable platform against the likelihood of moving - carefully compare the service offerings between different vendors and try to avoid using services with highly vendor-specific implementation requirements. Using self-contained services and containerising everything coupled with extensive test coverage will go a long way towards reducing this cost.